How Elon Musk Gambled Tesla to Save SolarCity | Vanity Fair

It was a Saturday night last March in Buffalo, and Dennis Scott was sitting at home. A stocky veteran with salt-and-pepper hair and a close-cropped beard, Scott had been laid off from Tesla’s factory in Buffalo two months earlier as part of a global reduction in the company’s workforce. Since then, he had taken to sending Elon Musk emails and point-blank tweets, describing the pain the layoffs were causing.

Ten days after Scott was let go, Musk had tweeted a goofy picture of himself posing with what looked like a machine gun. Scott retweeted the image and called Musk a clown. “If I were CEO and someone told me my company wasn’t working right,” he explains, “I wouldn’t be clowning around. I’ve got people counting on me for their livelihood.”

Now, around 10 p.m., his phone rang. The call was from an unmarked number. Scott answered.

“It’s the clown,” the person at the other end informed him.

Scott, unfazed, figured that Musk must have gotten his number from the company. For the next 20 minutes, he recalls, he and his former employer had a civil conversation. “When are you going to fix your company?” Scott asked.

Musk was pleasant but offered no specifics about the Buffalo plant. Scott continued to ask frank questions. “You took $750 million from New York,” he told Musk, referring to the taxpayer money that the state handed Tesla as part of its Buffalo Billion program to revitalize upstate New York. “You gave us hope that you were going to do something.”

Musk’s responses left Scott unimpressed. “Musk is a nice guy when you talk to him,” he says. “But I think he’s full of shit. He’ll tell you whatever you want to hear.”

Musk, after initially telling Vanity Fair that he had “no record” of the call, now denies that it ever took place. In public, he doesn’t talk much about Tesla’s factory in Buffalo—a place he once, in better times, dubbed Gigafactory 2. Gigafactory 1, of course, is Tesla’s much-hyped futuristic electric car plant outside Reno. Gigafactory 2, which is shrouded in silence and secrets, was a controversial side venture: a high-stakes move to dominate America’s growing market for solar energy. Tesla bought the factory’s main tenant, SolarCity, for almost $5 billion in 2016. The plan, in true Muskian hyperbole, was to turn the plant in Buffalo into what was billed as the largest manufacturing facility of its kind in the Western Hemisphere. SolarCity would build 10,000 solar panels per day and install them on homes and businesses across the country. In the process, it would create 5,000 jobs in an area that very much needed them. “This is one of the poorest cities in the country,” Scott says. “You get a big company here, and it’s a big deal.”

From the outside, the sheer scale of the Buffalo plant sparkles with promise. At 1.2 million square feet, it stands at the point where the Buffalo River bends through the city. The building is gleaming white, as if to signify its freshness amid a landscape of abandoned grain elevators and sprawling, desolate steel mills. The area around the factory is hardscrabble working class; until SolarCity was built, people only drove through it when the fierce wind off Lake Erie shut down the highway that residents take from the southern suburbs to downtown. Now three flags fly in front of the factory: those of the United States, New York State, and Tesla.

But three years after Tesla bought SolarCity, there are serious doubts as to whether the plant will ever fulfill its promises. The website CleanTechnica, which is mostly supportive of Musk, calls SolarCity “a disaster waiting to happen.” A potentially costly lawsuit alleges that Tesla acquired SolarCity at the expense of its own shareholders. And former employees want to know what happened to the massive subsidy Tesla received. “New York State taxpayers deserved more from a $750 million investment,” a laid-off employee named Dale Witherell wrote to Senator Kirsten Gillibrand. “Tesla has done a tremendous job providing smoke and mirrors and empty promises to the area.”

There are growing questions about SolarCity’s product, too. Last week, Walmart sued Tesla for breach of contract due to “years of gross negligence,” claiming that solar panels installed at seven of its stores went up in flames, causing millions of dollars in damage. The lawsuit, citing Tesla’s “utter incompetence,” seeks to have the company remove rooftop panels it installed at more than 240 Walmart stores.

The controversy over SolarCity, which has dovetailed with questions about Musk’s mountain of debt and profit shortfalls, offers a window into the mind-set of America’s most outlandish and unpredictable CEO. Musk’s believers argue that the details of his ventures don’t matter: It’s the grand vision that counts. “The guy has a will to make stuff happen that is extraordinary,” says someone who worked closely with Musk. “He willed Tesla to happen. And in willing a reality into existence, he might not stick to the facts.” But in the case of SolarCity, Musk’s penchant for making promises he can’t deliver on turned out to matter a great deal—and could even pose a threat to his entire empire.

When Witherell got his job at the SolarCity plant last year, he was thrilled. He’d moved back to Buffalo, where his parents live, after a stint in Texas and a tough divorce. He has a disabled daughter, but even so, the job wasn’t so much about the paycheck. “At some point push is going to come to shove in our world, and fossil fuel use is going to catch up with us,” he says. “I believed in the product.”

The plant, in fact, was a centerpiece of Governor Andrew Cuomo’s grand plan to revitalize upstate New York. Buffalo’s landscape offers a daily reminder of its past glory and present despair. The grain elevators that Le Corbusier once called the “magnificent first fruits of the new age”—and which supplied the nation for half a century, before they were rendered irrelevant by the St. Lawrence Seaway—still loom over the horizon, too costly to demolish. The blackened shell of the warehouse at the old Bethlehem Steel complex, destroyed by fire a few years ago, punctuates the river like an angry exclamation point. Once one of the nation’s largest cities, Buffalo’s empty streets feel oddly discordant. “Here, if you have to sit through a traffic light even once, you have an aneurysm,” says Dave Robinson, an editor at the Buffalo News.

Now, under Cuomo’s plan, Buffalo’s massive steel plants would be replaced with the sun itself. In September 2014, the governor toured the SolarCity site. The smiles were big and the words were grand. The success of the plant, Cuomo proclaimed, was “of critical importance to the United States’ economic competitiveness and energy independence.”

SolarCity was founded by two of Musk’s cousins, Lyndon and Peter Rive, who grew up with him in South Africa. Musk, who put in $10 million, was the largest shareholder and chairman of the board. The initial idea, the Rives explained, was not to be a manufacturer but rather to control the entire consumer experience of going solar, from sale to installation, thereby driving down costs. For a time, SolarCity was a hot stock, growing almost tenfold from its public offering in 2012 to its peak in early 2014.

As is common with Musk’s ventures, SolarCity professed to be focused on changing the world. “Everything was very motivational,” says a former executive. Some workers, taking the ethos to heart, sported SolarCity tattoos.

WHERE THE SUN DON’T SHINE
Tesla promised to create 1,460 jobs at its solar factory in Buffalo. It currently employs only 329.

Photograph by Tesla.

But the initial success of the company’s stock masked some difficult realities. SolarCity’s business model was to front the costs of installing solar panels and allow homeowners to pay over time, which created a constant need for cash. That required raising money from outside investors, often big banks, who were then entitled to the first chunk of the payments homeowners made—leaving SolarCity in a never-ending scramble to raise more debt. The real engineering that took place at SolarCity, in short, was financial, not environmental.

On the consumer side, SolarCity was plagued by complaints about misleading sales tactics and shoddy installations. As the problems mounted, some workers began to feel manipulated by the company’s talk about being a force for good in the world. “I turned a blind eye to a lot of the silliness because of the idealism,” says one former senior employee. “I don’t know when the Rubicon was crossed, but there were micro-crossings every day.”

By 2014, several insiders say, the board was also growing concerned. The company imported most of its solar panels from China, and it looked like demand would soon outpace supply. Because Musk had a reputation as a manufacturing genius, the board decided that SolarCity needed to start making its own panels—a huge shift in its business model. “Installing and selling solar has almost nothing to do with manufacturing,” says a former solar-industry executive. “It’s like a car dealer saying it’s going to make cars.”

In June 2014, SolarCity bought Silevo, a solar-panel manufacturer that had struck a deal with New York to build a factory in Buffalo. On a conference call, Musk boasted that the deal would enable SolarCity to install tens of gigawatts of panels every year—far beyond the company’s peak annual run rate of about one gigawatt. He spoke as if the technology were already proven. On its website, SolarCity predicted it would “achieve a breakthrough” in solar-power pricing thanks to “massive economies of scale.”

“It was shoot first and aim later,” says the former senior employee. “There was a lot of machismo going on: bigger, better, badder, faster.”

By the time Cuomo visited the site three months later, Silevo’s smallish deal had metastasized. The state promised to spend $350 million to build a factory and another $400 million on equipment specified by SolarCity. The company would get a 10-year lease on the facility—for just $1 a year. In return, it promised to employ at least 1,460 people in “high-tech” jobs at the factory, hire another 2,000 to support the sale and installation of solar panels in New York, and help attract an additional 1,440 “support jobs” in the state. Once it achieved full production, the company pledged, it would spend some $5 billion in New York over the following decade.

“It was sold as a perfect marriage,” says the former senior employee. “The area around the factory is terrible, and I remember thinking: Wow, we are going to save the town where steel was made.” Cuomo too was hooked. “He was enchanted with the idea of Elon Musk in Buffalo,” says a longtime lobbyist in Albany. “I think he actually thought Musk was the next Dalai Lama.”

Even then, to those who looked closely, the cracks at SolarCity were becoming apparent. In 2014, key executives had started to leave. The Rives began to sell stock. SolarCity’s debt was soaring, and the yield on its bonds hit double digits, a sign that the market thought the company was in trouble. Goldman Sachs, one of Musk’s major bankers, called SolarCity the “worst positioned” company for capitalizing on future growth in the solar sector. One of the few things shoring up the company’s stock, according to a former investor, were the constant rumors that Musk was somehow going to bail it out.

In reality, the situation was even uglier than outsiders knew. As SolarCity struggled to raise money from institutional investors, it began offering individuals a chance to buy what it called Solar Bonds. (“Now you can get paid while driving the solar revolution,” the marketing material said.) But there were few takers—so other parts of the Musk empire took up the slack. According to the shareholder lawsuit, SpaceX acquired $255 million of the bonds. Musk himself bought $75 million of them, and the Rives acquired another $38 million. To raise the cash, Musk borrowed against both Tesla and SolarCity stock, increasing his personal credit lines from $85 million to $475 million. He also used his own reputation to shore up the stock: In February 2016, when SolarCity stock plunged to its lowest level in three years, Musk bought $10 million in shares. A week later, when the news became public, the stock soared by almost 25 percent.

At the same time, according to the shareholder lawsuit against Tesla, the company faced “significant liquidity concerns”—meaning it was running out of money. An accounting inquiry from the SEC noted that SolarCity was burning through cash—$659 million in the first quarter of 2016 alone. That February, at a Tesla board meeting, Musk proposed a solution: Tesla, he said, should acquire SolarCity.

The board balked. But Musk kept pushing. Two weeks later, he proposed the acquisition again. Once again, the board said no.

It was a hopelessly conflicted situation. Musk owned more than 20 percent of both SolarCity and Tesla. His brother, Kimbal, served on both boards, as did several investors, including Antonio Gracias, a close friend of Musk’s. As a judge in the shareholder lawsuit ruled, it is “reasonably conceivable” that Musk effectively controlled the Tesla board when he pushed it to acquire SolarCity. (Tesla, which has dismissed the allegations in the lawsuit as false, insists that “all appropriate parties” recused themselves during the acquisition.)

At the time, Musk was still a heroic figure to many. As former Tesla board member Nancy Pfund once said, “He’s always been a master of the universe in my mind.” Even Tesla skeptics admit that the Model S, which was launched in 2012, will go down in history as an absolute classic, followed by the equally celebrated Model X in 2015. In those days, Tesla’s stock was trading at well over $200 a share, giving it a market value of more than $30 billion, a stunning figure for a company that hadn’t proven it could make money.

But over the years, many skeptics have come to see Musk’s stunts—from smoking pot during an interview to calling a diver who helped rescue kids trapped in a Thailand cave a “pedo guy”—as more unhinged than iconoclastic. One close observer of Musk recalls how he promised, back in 2001, to give away half of his equity in PayPal—dividing it evenly between “the people that have worked hard to build the company” and “causes I believe make the world a better place.” But Musk never made good on the pledge, and the observer came to see the episode as “symbolic of Musk’s penchant for making grandiose statements that he either knows are not true at the time he makes them, or that he has no real intent of following through on.” Others see Musk’s promises as purposefully manipulative. “Musk has a habit of overstating Tesla’s operational capabilities and its prospects for profitability, especially when the company is preparing to raise capital, collect customer deposits, or secure regulatory benefits,” says Brian Horey of Aurelian Partners, an investment firm.

Now the brewing problems at SolarCity threatened to give skeptics real ammunition against Musk—unless those problems could be buried. In May 2016, the Tesla board finally agreed to acquire the company for almost $5 billion, including the assumption of nearly $3 billion in SolarCity debt. On a conference call on June 22, the day after the deal was publicly announced, Musk told analysts and investors that the company had “the best technology out there for high-efficiency, low-cost solar panels.” He didn’t say anything about the liquidity crisis at SolarCity. Nor did he mention something else that shareholders allege the Tesla board came to learn as it did its due diligence on SolarCity: The cost per watt of solar modules being produced in Buffalo was actually projected to be 20 cents above the rest of the industry.

On October 28, 2016, just before shareholders were set to vote on the acquisition of SolarCity, Musk strode onto a platform erected on the set of Desperate Housewives at Universal Studios’ back lot in Los Angeles. He talked about the existential threat presented by global warming and the desperate need for sustainable energy. Then he gestured to a group of houses that had been set up around him. They might look normal, he said, but they actually featured a revolutionary new product called the Solar Roof—shingles that would last longer and cost less than a regular roof, even before factoring in electricity. Tesla expected production to begin the following summer.

The next month, shareholders approved Tesla’s acquisition of SolarCity. “Vote tally shows ~85% of unaffiliated shareholders in favor of the Tesla/SolarCity merger!” Musk tweeted. The deal doubled Tesla’s debt load, but it was good for Musk, who converted his stake in SolarCity into more than $500 million in Tesla stock. By preventing SolarCity from collapsing, he also shored up his most valuable asset: investor faith in his own genius. If any piece of his empire had faltered—if Musk were shown to be fallible rather than superhuman—it would have cast doubt on the narrative that enables him to raise cheap capital for his money-losing enterprises.

“Thanks for believing,” Musk tweeted to his shareholders.

That October, as Musk was making his pitch about the Solar Roof, a former Fortune 500 executive was watching it online at a friend’s barbecue. The former executive, who had spent years researching solar technology, understood what it took to make the Solar Roof work—and he was confident that Musk hadn’t figured it out. “He spewed total BS,” says the executive, who asked not to be identified. “I was flabbergasted. I was convinced in the moment that the shingles were fake.”

Adopting the Twitter handle @TeslaCharts, the executive began drawing on his Ph.D. in science, and his background as a financial analyst, to share infographics that detailed Musk’s overreach. His follower count mushroomed, and he became a core member of a group of outspoken Tesla critics who go by the Twitter hashtag #TSLAQ—Tesla’s stock symbol followed by the Q that companies pick up when they are delisted due to bankruptcy.

Many of them, in fact, were first drawn to Tesla by SolarCity, with its pile of debt and mountain of losses. “If it weren’t for SolarCity, #TSLAQ wouldn’t exist,” says @TeslaCharts. He points out that Musk faced a catch-22 of sorts: If he hadn’t bailed out SolarCity, his whole debt-laden empire might have cracked. Yet without the bailout, Tesla would be far more healthy. “When the history of Tesla is written,” he says, “the acquisition of SolarCity will be seen as the moment where the narrative took a decisive turn.”

Others shared @TeslaCharts’ suspicions about Solar Roof. Robinson, who covers SolarCity for the Buffalo News, had flown to Los Angeles for Musk’s presentation. Afterward, he asked an engineer for the company if the tiles Musk had pointed to were real. “Oh no,” the engineer replied. “These are dummies. We just popped them up here to show you.” Robinson wasn’t outraged—it made sense that Musk would show a prototype—but he took note of the contrast between the rhetoric and the reality. “They made it sound like they had figured out how to get it to work,” he says.

And Tesla continued to make it sound that way. In early 2018, the company announced that production of the Solar Roof had begun in Buffalo. That fall, Tesla told Bloomberg News it was “gearing up for tremendous growth in 2019. We have a product, we have the customers, we are just ramping it up to a point where it is sustainable.”

But in its quarterly letter, a month earlier, Tesla had confessed that the product wasn’t actually ready yet. “We continue to iterate,” the company wrote. In a legal filing, Tesla acknowledged that the much-hyped technology it had acquired from Silevo wasn’t all that it was cracked up to be. And last May, an investigation by Reuters revealed that most of the solar cells being produced in Buffalo were being sold overseas, not used in the Solar Roof, because demand was so low.

Customers who tried to purchase a Solar Roof took to Twitter to share their horror stories: Kevin Pereau, a California homeowner, said he paid a deposit of $2,000 to have a Solar Roof installed more than two years ago—then never heard from the company again. He got his money back only after he started tweeting at Musk every single day.

Musk, meanwhile, is still making promises. Last March, he proclaimed that 2019 would be the “year of the Solar Roof.” In late July, he tweeted that Tesla is “hoping” to turn out 1,000 Solar Roofs a week by the end of the year. But even onetime believers have become doubters. The MIT Technology Review, which included the Solar Roof in its list of 10 “breakthrough technologies” in 2016, now calls it a “flop.” In a recent analyst note, JP Morgan warned that Solar Roof will be a “niche” product at best. Musk has “sustained a kind of Kabuki theater in which the Solar Roof ramp is always imminent, but never here,” wrote investor John Engle, a #TSLAQ member.

Another #TSLAQ member, a Yale-trained lawyer and investment manager named Lawrence Fossi, made a discovery while combing through SolarCity’s financial statements. Without fanfare—and with no input from constituents—state officials had quietly issued a series of 10 amendments that watered down the requirements SolarCity must meet in exchange for the $1 lease on the Buffalo factory. The 1,460 “high-tech” jobs at the factory became just plain old jobs, as did the 2,000 jobs to support solar sales and installation in New York. The agreement to employ 900 people at the factory within two years shrank to 500. And the timing for the additional jobs was extended to 10 years after the factory was completed—at which point the lease would also be expiring. (Tesla argues that the company is now responsible for all 5,000 jobs, instead of being able to fulfill them through suppliers.) The governor’s office declined to comment on who authorized the changes, and state officials have yet to provide any public explanation as to why they opted to let a big corporation like Tesla off the hook.

In fact, the Buffalo deal turned out to have been tainted by corruption from the very start. Just one day after Tesla finalized its acquisition of SolarCity, Preet Bharara, then the U.S. attorney for the Southern District of New York, announced criminal charges against a handful of Cuomo staffers for rigging the construction bids for the Buffalo Billion program to favor the governor’s campaign donors. The man tapped by Cuomo to oversee the taxpayer subsidies, as well as a leading donor who received a $225 million contract to build out the Buffalo factory, were both convicted last year of conspiring to rig the bids

Lyndon and Peter Rive have both left SolarCity, and the company’s original business of installing solar roofs has all but evaporated. The company once controlled one-third of the residential market; now, according to the consulting firm Wood Mackenzie, its share is less than 7 percent. In the second quarter of this year, SolarCity installed only 29 megawatts of solar panels—far below the 10,000 megawatts in annual installations that Musk had promised. “Total implosion” is how one SolarCity insider describes it.

Across the street from the factory in Buffalo stands a small building that houses a coffee shop and an office space. Both were built to cater to the plant, says Robinson, the Buffalo News editor. The coffee shop is surviving, but the office space is vacant. What few jobs do exist at SolarCity barely compete with the local grocery store. “For $750 million, we’re getting jobs that pay $2 an hour more than Aldi’s,” says Robinson.

In a statement to Vanity Fair, Tesla argues that its jobs in Buffalo are competitive, especially when benefits and equity are factored in. It says it has expanded its operations at the factory to include “some of our most innovative and pioneering products.” And it accuses the magazine of presenting a “one-sided view with cherry-picked sourcing aimed at feeding into the fear, uncertainty, and doubt being circulated about Tesla every day by those looking to gain from Tesla’s losses.”

But the level of secrecy surrounding the SolarCity plant may offer an additional indication of how bad things are. Tesla refused to allow me to take a tour, and former employees say a rare media event at the factory last fall was highly scripted. “They spent more time and resources trying to fabricate what people saw than they do making anything,” says Witherell, who worked there at the time. “They told employees to pretend we were busy.” A story aired last February by News 4 Buffalo described the shop floor as “torpid,” with idle employees milling about. “They say they are in ‘ramp up’ mode,” says Scott, the former employee. “But this isn’t even start-up mode. What company spends two and a half years starting up something they were already supposed to be the best at?”

Last April, not long after he placed his late-night call to Scott, Elon Musk finally paid his first-ever visit to Buffalo. There was no press release, no triumphant post on social media, no meeting with reporters. Local authorities were in the midst of performing the genuine engineering feat of dismantling what’s known as the “ice boom”—hundreds of steel pontoons, spanning over a mile and a half, that keep the massive amounts of ice on Lake Erie from floating down the Niagara River and jamming up hydropower turbines. After the visit, Musk continued his upbeat assessment of production. “We are looking forward to scaling up significantly through the balance of this year and into next,” he said.

Musk’s other proclamations in recent months have been far grander. He has promised that by next year, Tesla will be producing self-driving cars—and deploying a fleet of 1 million robotaxis. He has claimed that Neuralink, his secretive start-up, has developed a “thread” that can be inserted into the human brain, merging our minds with artificial intelligence. And he is seeking approval to build an underground “hyperloop” that will whisk passengers between Washington, D.C., and Baltimore in 15 minutes.

When Tesla bought SolarCity, it said the deal would “add more than half a billion dollars in cash to Tesla’s balance sheet over the next three years.” But it appears to have had the opposite effect. “I think it’s a big source of the cash-flow deficit,” says one longtime analyst. “I think it is a big thorn inside of Tesla.” The company has paid back some of SolarCity’s debt, including the Solar Bonds owed to Musk and SpaceX. But this fall, another $556 million is coming due. In a characteristic tweet, Musk once vowed he would “personally” repay the SolarCity debt if need be.

There may be another cost. By next April, Tesla is required to start paying an annual fine of $41.2 million if it fails to employ 1,460 people in Buffalo. Tesla says it currently has 636 employees statewide in New York, including 329 at the plant, and that it has invested almost $400 million in New York. Engle, the TSLAQ member, argues that Tesla cannot afford to admit that SolarCity has been a fiasco, because doing so would open the company to significant liability in the ongoing lawsuit over the acquisition.

Officials in New York, meanwhile, appear to be taking belated steps to document what is really happening in Buffalo. Last spring, the state announced that it was auditing all of its high-tech programs, with a focus on Tesla. Everyone in Albany, says the longtime lobbyist, has accepted that the Buffalo plant is a “disaster”—a poster child for why government giveaways to big companies don’t work.

But the official who took credit for the deal with Tesla—the man who championed the company as a Rust Belt savior—stands by his decision to place his trust in Elon Musk. Governor Cuomo, who paid his own visit to Buffalo last spring, declared that he’s perfectly pleased with the progress at SolarCity. “They’re ahead of schedule,” he said.

This post has been updated.

Corrections: The 1,460 jobs Tesla must provide by April 2020 can be anywhere in Buffalo, not simply at the Buffalo plant. In addition, Tesla’s quarterly letter acknowledging that it was continuing to “iterate” on the Solar Roof was issued in October 2018, not several months later. And SolarCity once controlled one-third of the residential solar market, not two-thirds.